Although Convolutional Neural Networks provided effective solutions for computer vision tasks across various domains in recent years, the trained times of the models remain a crucial aspect. A straightforward way to decrease the computational costs is the usage of smaller images for training. However, existing image resizing approaches are either computationally intensive or lossy methods. This paper presents the usage of the Cantor pairing function for image resizing through feature reduction. The pairing function is a fast bijective transformation that maps a pair of two integers into a single value. Combining two pixels into one leads to a decrease in the image size, while the bijective property ensures a lossless transformation. The experiments conducted on the MNIST dataset demonstrated that the proposed approach reduces resizing time at least 6 times, decreases training times significantly, and achieves model performance at least as good as the baseline methods.
